A comparison of farm typology approaches in Northern Ghana

K. S. Kuivanen, S. Alvarez, M. Michalscheck, S. Adjei-Nsiah, K. Descheemaeker, S. Bedi Mellon, J.C.J. Groot

Introduction Variables/ Criteria

Objective

Comparison of two approaches to the typification of 80 surveyed farm

households across three Africa RISING project communities in

Ghana’s Northern Region: statistical (top-down, researcher-defined)

and participatory (community-based, farmer-defined).

Conclusions

Figure 1. Framework of the statistical- and participatory approaches to typification (PCA:

Principal component analysis; CA: Cluster analysis; PLA: Participatory Learning and Action).

Acknowledgements We wish to thank the participating experts and farmers. The case study was financed by the

Africa RISING program funded by the Feed the Future Initiative of the United States Agency for

International Development (USAID). Survey data collection and participatory research were

coordinated by the International Institute of Tropical Agriculture (IITA) in Tamale, Ghana.

• Types D and E in participatory typology could be important target

groups in the R4D project, but were not represented in statistical

typology.

• For both typologies data were sometimes inaccurate: for the

statistical typology survey data did not always reflect the reality (for

various reasons), for the participatory typology cultural and social

(power) issues tended to distort assessment of farms.

• Limited overlap between assignment of surveyed households to

types when comparing the two typologies.

• Nevertheless, mean profile of participatory types roughly matches

that of corresponding statistical types on selected variables.

(Dis)similarity (Figure 2.)

• Statistical: variables combined into a smaller number of dimensions

(PCA) before clustering into six types (CA).

• Participatory: criteria used sequentially to group farmers according

to farm size and then subdividing classes on the basis of other

relevant criteria into five types.

• The statistical typology revealed general structure/ pattern of farm

household variation: possibility for extrapolation beyond study area.

• By incorporating actor perspectives, the participatory typology

captured context-specific aspects of farm complexity: may enhance

local relevance and socio-cultural sensitivity of interventions.

Methods

Grouping diverse farm households into subsets or ideal types can

support the development (selection of farms), implementation

(targeting and scaling-out of innovations) and monitoring (scaling up

of impact assessments) of R4D projects. Different approaches to

typology construction can have different results and this will affect the

meaningfulness of the resulting types for involved stakeholders.

Results

• Selection differed between statistical (12 quantitative key variables

comprising household size, labour, land use, livestock and income)

and participatory groupings (15 criteria - farm size and gender most

discriminating).

• Similar descriptive names belied divergent underlying meaning due

to interpretation/cultural differences.

Type delineation (Table 1.)

Figure 2. Classificatory overlap between the typologies (A) and kernel density curves

(participatory Types A-C, group means represented by coloured dashed lines) combined with

boxplots (statistical Types 1-6, group means represented by black markers) for the variable of

herd size (total TLU) per Type (B).

Type Symbol Main characteristics % in

survey

Statistical Typology

1 HRE, large cattle herd, ample off-farm activities 11%

2 MRE, large farms, market orientation 10%

3 MRE, small ruminants, on-farm labour intensive 13%

4 MRE, small ruminants, ample hired labour 46%

5 LRE, maize dominated, few off-farm activities 14%

6 SRC, livestock sales, ample off-farm activities 6%

Participatory typology

A Pukparkara (‘Big farmers’- men): HRE, market-orientation 8%

B Pukparsagsa (‘Medium farmers’- men): MRE, variable orientation

52%

C Pukparbihi (‘Small farmers’- men): LRE, subsistence orientation

40%

D Pagba pubihi (‘Small farmers’- women & children): LRE and SRC, market orientation

0%

E Suhukpion (‘Farm-less’- men): work on other farms as hired labour

0%

Table 1. Main characteristics of types (HRE: High resource endowed; MRE: Medium resource

endowed; LRE: Low resource endowed; SRC: Severely resource constrained)

Farmer symbols: Type A. Household heads are always happy and smiling; Type B. Fist and outstretched hand indicate that what these farmers have is not enough, they need more to be self-sufficient; Type C. The hoe symbolizes that the farmers cannot afford to hire the services of a tractor; Type D. The cooking pot and cutlass are tools used by women; Type E. The ear suggests that the farm-less always listen out for work opportunities.

Wageningen University

P.O. Box 430, 6700 AK Wageningen

Contact: jeroen.groot@wur.nl

T + 31 (0)317 48 59 24

www.wageningenUR.nl/fse